Feasibility of predicting Ki-67 expression in breast cancer using radiomics nomogram based on magnetic resonance diffusion weighted imaging
10.3969/j.issn.1005-202X.2025.07.013
- VernacularTitle:基于磁共振弥散加权成像的影像组学列线图预测乳腺癌Ki-67表达的可行性
- Author:
Junli XU
1
;
Xueyuan TAN
1
;
Zhiling WEN
1
;
Yudi LIANG
1
Author Information
1. 广东医科大学附属第二医院放射影像科,广东 湛江 524000
- Publication Type:Journal Article
- Keywords:
breast cancer;
magnetic resonance imaging;
diffusion weighted imaging;
apparent diffusion coefficient;
Ki-67 expression;
nomogram
- From:
Chinese Journal of Medical Physics
2025;42(7):929-934
- CountryChina
- Language:Chinese
-
Abstract:
Objective To explore the feasibility of radiomics nomogram based on magnetic resonance diffusion weighted imaging for predicting the expression of Ki-67 in breast cancer.Methods A retrospective study was conducted on patients with breast cancer confirmed by surgery and pathology in the Second Affiliated Hospital of Guangdong Medical University.All patients were detected by Ki-67 expression staining,and then divided into group A(n=28,low-level expression of Ki-67)and group B(n=73,high-level expression of Ki-67).The apparent diffusion coefficient(ADC)graph was generated from diffusion weighted images,and the two groups were compared for radiomics features of ADC images and clinical data.The expression level of Ki-67 in breast cancer was predicted using the features of LASSO after dimensionality reduction,and a nomogram model was established,whose diagnostic efficiency was analyzed with receiver operating characteristic curve.Results No significant difference was observed in ADC value,age,carbohydrate antigen 199,carbohydrate antigen 153,carbohydrate antigen 125 and carcinoembryonic antigen between two groups(P>0.05).LASSO regression identified two radiomics features as predictors for the expression level of Ki-67 in breast cancer.The best tuning Lambda of LASSO was-0.125 690 135 478 682,and the included radiomics features for nomogram establishment were MinIntensity and Quantile95.The established nomogram had an area under ROC curve of 0.917,achieving a sensitivity of 91.7%and a specificity of 83.3%.Conclusion The expression of Ki-67 in breast cancer can be predicted based on the radiomics features of ADC images,which can provide a noninvasive detection method for evaluating the proliferation degree and treatment prognosis of breast cancer.
